ABSTRACT Autism spectrum disorder (ASD) is a complex neurodevelopmental condition with atypical social communication as a core symptom. Variations in social information processing in individuals with ASD are associated with the social brain, which encompasses four specific subnetworks, that is, reward system, theory of mind network, mirror neuron system, and face perception network. However, the relationship between neural mechanisms of altered social functioning and modular integration of these subnetworks within the social brain remains unclear in ASD. With resting‐state functional MRI (rs‐fMRI) data from two large‐scale datasets (ABIDE I and II), we computed the participation coefficient to explore the abnormal modular integration of the four subnetworks in 298 ASDs and 348 typically developing (TD) controls. Then, its associations with clinical symptoms, neurotransmitter systems, and transcriptional signatures were investigated. Additionally, the age effect on aberrant modular integration was estimated with linear regression models. Finally, we assessed the reproducibility of our results from a meta‐perspective using other datasets. ASD participants exhibited increased integration of the reward system relative to TDs, which was correlated with Social Responsiveness Scale total score, the neurotransmitters such as 5HT1a and GABAa, and the disruption of the transcriptional signatures including cell proliferation and migration as well as tube and tissue morphogenesis. Additionally, the modular integration abnormality of the reward system was stable across development and replicated across datasets. We revealed a symptom‐related, neurotransmitter‐ and transcriptional signature‐associated, age‐stable, and reproducible modular integration abnormality of the reward system in ASD. This hyper‐integration was linked to reduced GABAa and serotonin receptor densities, providing neuroimaging and molecular evidence supporting the excitatory‐inhibitory imbalance theory of ASD and insights into the mechanisms underlying social variations in ASD.
Yang et al. (Mon,) studied this question.